基于在线压缩重构的非侵入式电力负荷监测

doi:10.13229/j.cnki.jdxbgxb.20220927

摘要/Abstract

摘要：

对家庭场景下电力负荷的精确监测常依赖复杂庞大的算法模型，难以在边缘设备中部署。同时，海量电力数据给电力网络通信带来了巨大的挑战。本文针对以上问题，提出了一种分布式非侵入式电力负荷监测方法。通过基于注意力机制的长短时记忆网络（LSTM）负荷监测算法计算并识别负荷设备运行状态，借助云边协同技术将负荷监测任务分布式部署于云端以及边缘端中，解决边缘算力不足的问题。针对云边通信带来的高网络带宽需求，通过基于K奇异值分解（K-SVD）双稀疏在线字典学习的压缩感知方法对负荷信号进行压缩和重构，有效缓解通信资源紧张。对比不同负荷场景下监测算法的表现，结果表明：本文负荷监测算法针对不同的负荷类型均可以保持95%以上的准确率。设计实验验证了本文压缩感知方法对电力负荷信号压缩的有效性，确定负荷数据无失真压缩感知最大压缩比。

关键词: 通信与信息系统, 非侵入式负荷监测, 压缩感知, 分布式部署, 长短时记忆网络

Abstract:

Accurate monitoring of power loads in home scenarios often relies on complex algorithmic models that are difficult to deploy in edge devices. At the same time， massive power data poses a huge challenge to communication of grid. In response to the above issues， this paper proposes a distributed non-intrusive load monitoring method. This method calculates and identifies the operating state of the load by the LSTM load monitoring algorithm based on the attention mechanism， and distributes the load monitoring task in the cloud and the edge with the help of cloud-edge collaboration technology to solve the problem of insufficient edge computing power. Aiming at the high network bandwidth requirements brought about by cloud-side communication， the compressed sensing method based on K-SVD double sparse online dictionary is used to compress and reconstruct the load signal， which effectively alleviates the shortage of communication resources. Comparing the performance of the monitoring algorithm under different load scenarios， the results show that the load monitoring algorithm in this paper can maintain an accuracy rate of more than 95%. Experiments are conducted to verify the effectiveness of the compressed sensing method on the compression of load signals， and determine the maximum compression ratio of load data without distortion.

Key words: communication and information system, non-intrusive load monitoring, compressed sensing, distributed deployment, long short-term memory

中图分类号:

TP274

周求湛,冀泽宇,王聪,荣静. 基于在线压缩重构的非侵入式电力负荷监测[J]. 吉林大学学报(工学版), 2024, 54(6): 1796-1806.

Qiu-zhan ZHOU,Ze-yu JI,Cong WANG,Jing RONG. Non-intrusive load monitoring via online compression and reconstruction[J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(6): 1796-1806.

图/表 12

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

表1

图11

参考文献 42

1	Gillis J M, Chung J A, Morsi W G. Designing new orthogonal high-order wavelets for nonintrusive load monitoring[J]. IEEE Transactions on Industrial Electronics, 2018, 65(3): 2578-2589.
2	Wang A L, Chen B X, Wang C G, et al. Non-intrusive load monitoring algorithm based on features of V⁃I trajectory[J]. Electric Power Systems Research, 2018, 157: 134-144.
3	Hart G W. Nonintrusive appliance load monitoring[J]. Proceedings of the IEEE, 1992, 80(12): 1870-1891.
4	张露. 非侵入式负荷监测算法研究[D].广州: 华南理工大学电力学院, 2019.
	Zhang Lu. Research on non-intrusive load monitoring algorithms[D]. Guangzhou: School of Electrial Power Engineering, South China University of Technology, 2019
5	刘博. 非侵入式电力负荷监测与分解技术[D].天津:天津大学电气与自动化工程学院, 2014.
	Liu Bo. Non-intrusive power load monitoring and disaggregation technique[D].Tianjin: School of Electrial Information Engineering, Tianjin University, 2014
6	Hassan T, Javed F, Arshad N. An empirical investigation of V-I trajectory based load signatures for non-intrusive load monitoring[J]. IEEE Transactions on Smart Grid, 2014, 5(2): 870-878.
7	Tsai M S, Lin Y H. Modern development of an adaptive non-intrusive appliance load monitoring system in electricity energy conservation[J]. Applied Energy, 2012, 96: 55-73.
8	Batra N, Kelly J, Parson O, et al. NILMTK: an open source toolkit for non-intrusive load monitoring[C]∥e-Energy 2014-Proceedings of the 5th ACM International Conference on Future Energy Systems.New York,USA : Association for Computing Machinery, 2014: 265-276.
9	王轲, 钟海旺, 余南鹏, 等. 基于seq2seq和Attention机制的居民用户非侵入式负荷分解[J] .中国电机工程学报, 2019, 39(1): 75-83, 322.
	Wang Ke, Zhong Hai-wang, Yu Nan-peng, et al. Nonintrusive load monitoring based on sequence-to- sequence model with attention mechanism[J]. Proceedings of the CSEE, 2019, 39(1): 75-83, 322.
10	Yue Z, Witzig C R, Jorde D, et al. BERT4NILM: a bidirectional transformer model for non-intrusive load monitoring[C]∥NILM 2020-Proceedings of the 5th International Workshop on Non-Intrusive Load Monitoring.New York,USA :Association for Computing Machinery, 2020: 89-93.
11	凌家源. 基于事件检测与深度学习的非侵入式负荷监测算法[D]. 杭州: 浙江大学电气工程学院, 2021.
	Ling Jia-yuan. Non-intrusive load monitoring based on event detection and deep learning[D].Hangzhou: College of Electrical Engineering, Zhejiang University, 2021.
12	Sun Y, Cui C, Lu J, et al. Data compression and reconstruction of smart grid customers based on compressed sensing theory[J]. International Journal of Electrical Power and Energy Systems, 2016, 83: 21-25.
13	Saavedra E, Mascaraque L, Calderon G, et al. The smart meter challenge: feasibility of autonomous indoor iot devices depending on its energy harvesting source and iot wireless technology[J]. Sensors, 2021, 21(22).
14	周学斌. 智能电网海量数据轻型化方法研究[D].武汉: 武汉大学电气与自动化学院, 2020.
	Zhou Xue-bin. Research on the method of light massive data in smart grid[D].Wuhan: School of Electrical Engineering and Automation, Wuhan University, 2020.
15	王大伟, 崔国宇, 庞振江, 等. 边缘计算在多芯模组化电能表中的应用研究[J]. 单片机与嵌入式系统应用, 2019, 19(4): 30-33, 37.
	Wang Da-wei, Cui Guo-yu, Pang Zhen-jiang, et al. Application research of edge computing in multi-core modular energy meter[J]. Microcontrollers & Embedded Systems, 2019, 19(4): 30-33, 37.
16	Donoho D L. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289-1306.
17	Candes E J, Tao T. Near-optimal signal recovery from random projections: universal encoding strategies?[J]. IEEE Transactions on Information Theory, 2006, 52(12): 5406-5425.
18	Bao K, Ibrahimov K, Wagner M, et al. Enhancing neural non-intrusive load monitoring with generative adversarial networks[J]. Energy Informatics, 2018, 1(): 295-302.
19	Linh N V, Arboleya P. Deep learning application to non-intrusive load monitoring[J]. 2019 IEEE Milan PowerTech, 2019, 2019: 2-6.
20	Vinay M B, Rekha K S. A model of saliency-based visual attention for rapid scene analysis[J]. International Journal of Recent Technology and Engineering, 2019, 7(6): 412-415.
21	Liu G, Guo J. Bidirectional LSTM with attention mechanism and convolutional layer for text classification[J]. Neurocomputing, 2019, 337: 325-338.
22	Woo S, Park J, Lee J Y, et al. Cbam: Convolutional block attention module[C]∥Proceedings of the European conference on computer vision (ECCV),Copenhagen, Denmark, 2018: 3-19.
23	Rubinstein R, Zibulevsky M, Elad M. Double sparsity: learning sparse dictionaries for sparse signal approximation[J]. IEEE Transactions on Signal Processing, 2010, 58(3): 1553-1564.
24	Mazumder R, Friedman J H, Hastie T. SparseNet: coordinate descent with nonconvex penalties[J]. Journal of the American Statistical Association, 2011, 106(495): 1125-1138.
25	Engan K, Aase S O, Husoy J H. Method of optimal directions for frame design[J]∥IEEE International Conference on Acoustics, Speech and Signal Processing-Proceedings, Piscetawar NJ: IEEE, 1999, 5: 2443-2446.
26	Aharon M, Elad M, Bruckstein A. K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation[J]. IEEE Transactions on Signal Processing, 2006, 54(11): 4311-4322.
27	Lu C, Shi J, Jia J. Online robust dictionary learning[C]∥Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition,Portland,USA, 2013: 415-422.
28	Mairal J, Bach F, Ponce J, et al. Online learning for matrix factorization and sparse coding[J]. Journal of Machine Learning Research, 2010, 11: 19-60.
29	Labusch K, Barth E, Martinetz T. Robust and fast learning of sparse codes with stochastic gradient descent[J]. IEEE Journal on Selected Topics in Signal Processing, 2011, 5(5): 1048-1060.
30	Sulam J, Ophir B, Zibulevsky M, et al. Trainlets: dictionary learning in high dimensions[J]. IEEE Transactions on Signal Processing, 2016, 64(12): 3180-3193.
31	Huang K, Wu Y, Long C, et al. Adaptive process monitoring via online dictionary learning and its industrial application[J]. ISA Transactions, 2021, 114: 399-412.
32	Tropp J A, Gilbert A C. Signal recovery from random measurements via orthogonal matching pursuit[J]. IEEE Transactions on Information Theory, 2007, 53(12): 4655-4666.
33	Blumensath T, Davies M E. Iterative thresholding for saparse approximations[J]. Journal of Fourier Analysis and Applications, 2008, 14(5/6): 629-654.
34	Baraniuk R, Davenport M, DeVore R, et al. A simple proof of the restricted isometry property for random matrices[J]. Constructive Approximation, 2008, 28(3): 253-263.
35	Baraniuk R G. Compressive sensing[J]. IEEE Signal Processing Magazine, 2007, 24(4): 118-121.
36	Candes E J, Tao T. Decoding by linear programming[J]. IEEE Transactions on Information Theory, 2005, 51(12): 4203-4215.
37	Donoho D L. For most large underdetermined systems of linear equations the minimal ℓ₁-norm solution is also the sparsest solution[J]. Communications on Pure and Applied Mathematics, 2006, 59(6): 797-829.
38	Needell D, Vershynin R. Uniform uncertainty principle and signal recovery via regularized orthogonal matching pursuit[J]. Foundations of Computational Mathematics, 2009, 9(3): 317-334.
39	石光明, 刘丹华, 高大化, 等. 压缩感知理论及其研究进展[J]. 电子学报, 2009, 37(5): 1070-1081.
	Shi Guang-ming, Liu Dan-hua, Gao Da-hua, et al. Advances in theory and application of compressed sensing[J]. Acta Electronica Sinica, 2009, 37(5): 1070-1081.
40	Kelly J, Knottenbelt W. The UK-DALE dataset, domestic appliance-level electricity demand and whole-house demand from five UK homes[J]. Scientific Data, 2015, 2: 1-14.
41	Shalev-Shwartz S. Online learning and online convex optimization[J]. Foundations and Trends in Machine Learning, 2011, 4(2): 107-194.
42	Hoyer P O. Non-negative matrix factorization with sparseness constraints[J]. Journal of Machine Learning Research, 2004, 5: 1457-1469.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

在线学习方法	稀疏度	执行时间/s
ODL	0.9498	101.34
K-SVD双稀疏在线字典学习	0.9541	0.87

[1]	吴飞,农皓业,马晨浩. 基于粒子群优化算法⁃长短时记忆模型的刀具磨损预测方法[J]. 吉林大学学报(工学版), 2023, 53(4): 989-997.
[2]	周求湛,冀泽宇,王聪,胡继康,李明明,陈禹竺,周险峰,刘萍萍. 基于猫群算法的震动感知周界安防系统[J]. 吉林大学学报(工学版), 2023, 53(4): 1187-1199.
[3]	刘嫣然,孟庆瑜,郭洪艳,李嘉霖. 图注意力模式下融合高精地图的周车轨迹预测[J]. 吉林大学学报(工学版), 2023, 53(3): 792-801.
[4]	曾春艳,严康,王志锋,王正辉. 多尺度生成对抗网络下图像压缩感知重建算法[J]. 吉林大学学报(工学版), 2023, 53(10): 2923-2931.
[5]	孙洪亮,沈伟达,陈玲玲. 时延QoS约束下的混合业务带宽补偿算法[J]. 吉林大学学报(工学版), 2022, 52(8): 1912-1917.
[6]	陈建,于帆,林琳,孙铭会. 基于多阵列合成孔径的局部超声阵列聚焦方法[J]. 吉林大学学报(工学版), 2022, 52(10): 2447-2455.
[7]	窦慧晶,丁钢,高佳,梁霄. 基于压缩感知理论的宽带信号波达方向估计[J]. 吉林大学学报(工学版), 2021, 51(6): 2237-2245.
[8]	刘洲洲,张倩昀,马新华,彭寒. 基于优化离散差分进化算法的压缩感知信号重构[J]. 吉林大学学报(工学版), 2021, 51(6): 2246-2252.
[9]	侯春萍,赵春月,王致芃,田海瑞. 基于有效异常样本构造的视频异常检测算法[J]. 吉林大学学报(工学版), 2021, 51(5): 1823-1829.
[10]	王义君,张有旭,缪瑞新,豆佳敏. 5G中基于系统中断概率的D2D资源分配算法[J]. 吉林大学学报(工学版), 2021, 51(1): 331-339.
[11]	金心宇,谢慕寒,孙斌. 基于半张量积压缩感知的粮情信息采集[J]. 吉林大学学报(工学版), 2021, 51(1): 379-385.
[12]	胡钊政,李招康,陶倩文. 基于分布式二维激光测距仪的室内行人检测与跟踪[J]. 吉林大学学报(工学版), 2020, 50(2): 719-729.
[13]	赵鹏,蒋宇中,陈斌,李春腾,张杨勇. 基于局部方差域自适应Blanking的超低频信道噪声抑制方法[J]. 吉林大学学报(工学版), 2019, 49(5): 1696-1705.
[14]	单泽彪,刘小松,史红伟,王春阳,石要武. 动态压缩感知波达方向跟踪算法[J]. 吉林大学学报(工学版), 2018, 48(6): 1938-1944.
[15]	刘洲洲, 王福豹. 改进的离散混合蛙跳算法压缩感知信号重构及应用[J]. 吉林大学学报(工学版), 2016, 46(4): 1261-1268.